nV News Forums

 
 

nV News Forums (http://www.nvnews.net/vbulletin/index.php)
-   NVIDIA Linux (http://www.nvnews.net/vbulletin/forumdisplay.php?f=14)
-   -   1.0-8762 results in "black screen", 1.0-7676 does not. Using FX5200. (http://www.nvnews.net/vbulletin/showthread.php?t=72766)

alaric 07-01-06 12:59 PM

1.0-8762 results in "black screen", 1.0-7676 does not. Using FX5200.
 
1 Attachment(s)
Hiya, I've done a lot of reading and testing in the past days.
In the beginning, both my monitors (TwinView configuration) go blank and turn off,
as soon as I switch to X (startx), after upgrading the Nvidia drivers from 1.0-7676
to 1.0-8762.

I've tested this with three different kernel versions (all fedora core 4 packages):
2.6.14-1.1644_FC4, 2.6.16-1.2111_FC4 and 2.6.17-1.2139_FC4.
All of them give the same symptoms.

I've upgraded my A7V BIOS from 1003 to 1012.01a without any differences.

Normally I use Option "NvAGP" "1" (this was needed in the past, when I was using
a custom kernel), though I now see these kernels already have an AGPPart compiled
in - O well. I guess they both work now. Nevertheless, for these tests - and in the bug
report included below - I've already set Option "NvAGP" "0", to rule out AGP problems.

Looking at the bug report, it seems that 1.0-8762 has problems calculation the right
parameters for the monitors to use. I don't know why 1.0-7676 doesn't have problems
with that (if you want, I can post a "bug report" of a successful start with the older drivers,
to compare).

I've already reduced the hor and ver freq ranges a bit, from what normally works, but
that didn't make a difference either.

I'm pretty sure that with the right options in the xorg.conf I should be able to use the newer
version (what I need, because I want to upgrade my kernel, which is needed for me to be
able to burn DVD's again, which is badly needed cause I'm running out of diskspace...).
Searching this forum, I saw several suggestions - some of which I tried. Doing this, my
left monitor started to work! The right monitor still remained blank. For the bug report I have
removed those experimental options again, but - to my surprise - the left monitor continued
to work?!

I'd greatly appreciate some help here. I'm fairly confident that once both monitors work,
that it will continue to work with NvAGP to 3; so I suppose I'll keep it at 0 for now.

Thanks, Carlo

alaric 07-01-06 05:40 PM

Re: 1.0-8762 results in "black screen", 1.0-7676 does not. Using FX5200.
 
I did some more tests, and this is _definitely_ a bug in the new driver.

This is what I get with 1.0-7676:

[code]
(II) NVIDIA(0): --- EDID Information for display device CRT-0 ---
(II) NVIDIA(0): Manufacturer: IVM Model: 1730 Serial#: 18023
(II) NVIDIA(0): Year: 2006 Week: 1
(II) NVIDIA(0): EDID Version: 1.0
(II) NVIDIA(0): Analog Display Input, Input Voltage Level: 0.714/0.286 V
(II) NVIDIA(0): Signal levels configurable
(II) NVIDIA(0): Sync: Separate Composite SyncOnGreenSerration on. V.Sync Pulse req. if CompSync or SyncOnGreen
(II) NVIDIA(0): Max H-Image Size [cm]: horiz.: 32 vert.: 24
(II) NVIDIA(0): Gamma: 2.53
(II) NVIDIA(0): DPMS capabilities: StandBy Suspend; RGB/Color Display
(II) NVIDIA(0): redX: 0.625 redY: 0.340 greenX: 0.290 greenY: 0.605
(II) NVIDIA(0): blueX: 0.150 blueY: 0.246 whiteX: 0.783 whiteY: 0.297
(II) NVIDIA(0): Supported VESA Video Modes:
(II) NVIDIA(0): 720x400@70Hz
(II) NVIDIA(0): 720x400@88Hz
(II) NVIDIA(0): 640x480@60Hz
(II) NVIDIA(0): 640x480@67Hz
(II) NVIDIA(0): 640x480@72Hz
(II) NVIDIA(0): 640x480@75Hz
(II) NVIDIA(0): 800x600@56Hz
(II) NVIDIA(0): 800x600@60Hz
(II) NVIDIA(0): 800x600@72Hz
(II) NVIDIA(0): 800x600@75Hz
(II) NVIDIA(0): 832x624@75Hz
(II) NVIDIA(0): 1024x768@87Hz (interlaced)
(II) NVIDIA(0): 1024x768@60Hz
(II) NVIDIA(0): 1024x768@70Hz
(II) NVIDIA(0): 1024x768@75Hz
(II) NVIDIA(0): 1280x1024@75Hz
(II) NVIDIA(0): 1152x870@75Hz
(II) NVIDIA(0): Manufacturer's mask: 0
(II) NVIDIA(0): Supported Future Video Modes:
(II) NVIDIA(0): #0: hsize: 1280 vsize 1024 refresh: 60 vid: 32897
(II) NVIDIA(0): #1: hsize: 1280 vsize 1024 refresh: 74 vid: 36481
(II) NVIDIA(0): Supported additional Video Mode:
(II) NVIDIA(0): clock: 107.2 MHz Image Size: 300 x 225 mm
(II) NVIDIA(0): h_active: 1280 h_sync: 1322 h_sync_end 1426 h_blank_end 1667 h_border: 0
(II) NVIDIA(0): v_active: 1024 v_sync: 1030 v_sync_end 1033 v_blanking: 1065 v_border: 0
(II) NVIDIA(0): Supported additional Video Mode:
(II) NVIDIA(0): clock: 135.0 MHz Image Size: 300 x 225 mm
(II) NVIDIA(0): h_active: 1280 h_sync: 1312 h_sync_end 1456 h_blank_end 1712 h_border: 0
(II) NVIDIA(0): v_active: 1024 v_sync: 1024 v_sync_end 1027 v_blanking: 1064 v_border: 0
(II) NVIDIA(0):
(II) NVIDIA(0): EDID (in hex): 02000c003c171b6c62dc037176640000
(II) NVIDIA(0): 10011000f30281998c40880a75a4b962
(II) NVIDIA(0): f38cc4ffff08180818e8000000000000
(II) NVIDIA(0): 0000000000000e92003815009204a286
(II) NVIDIA(0): 3600c21e01000040cb43000b15008204
(II) NVIDIA(0): 02093000c21e01000081000000000000
(II) NVIDIA(0): 00000000000000000000000000000000
(II) NVIDIA(0): 0000000000000000000000000000006d
(II) NVIDIA(0): 02000c003c171b6c62dc037176640000
(II) NVIDIA(0): 10011000f30281998c40880a75a4b962
(II) NVIDIA(0): f38cc4ffff08180818e8000000000000
(II) NVIDIA(0): 0000000000000e92003815009204a286
(II) NVIDIA(0): 3600c21e01000040cb43000b15008204
(II) NVIDIA(0): 02093000c21e01000081000000000000
(II) NVIDIA(0): 00000000000000000000000000000000
(II) NVIDIA(0): 0000000000000000000000000000006d
(II) NVIDIA(0):
(II) NVIDIA(0): --- End of EDID Information for display device CRT-0 ---


This is what I get with 1.0-8762 (no other changes):

(--) NVIDIA(0): --- EDID for P@@ &?0?gF (CRT-0) ---
(--) NVIDIA(0): EDID Version : 2.0
(--) NVIDIA(0): Manufacturer : P@@
(--) NVIDIA(0): Monitor Name : P@@ &?0?gF
(--) NVIDIA(0): Product ID : 29123
(--) NVIDIA(0): 32-bit Serial Number : 0
(--) NVIDIA(0): Serial Number String :
(--) NVIDIA(0): Manufacture Date : 52518, week 177
(--) NVIDIA(0): DPMS Capabilities :
(--) NVIDIA(0): Prefer first detailed timing : No
(--) NVIDIA(0): Supports GTF : No
(--) NVIDIA(0): Maximum Image Size : 0mm x 0mm
(--) NVIDIA(0): Valid HSync Range : 4294967 kHz - 0 kHz
(--) NVIDIA(0): Valid VRefresh Range : -1 Hz - 0 Hz
(--) NVIDIA(0):
(--) NVIDIA(0):
(--) NVIDIA(0):
(--) NVIDIA(0): Raw EDID bytes:
(--) NVIDIA(0):
(--) NVIDIA(0): 20 00 c0 00 c3 71 b1 c6 26 cd 30 17 67 46 00 00
(--) NVIDIA(0): 01 10 01 00 3f 20 18 99 c8 04 88 a0 57 4a 9b 26
(--) NVIDIA(0): 3f c8 4c ff ff 80 81 80 81 8e 00 00 00 00 00 00
(--) NVIDIA(0): 00 00 00 00 00 00 e0 29 00 83 51 00 29 40 2a 68
(--) NVIDIA(0): 63 00 2c e1 10 00 00 04 bc 34 00 b0 51 00 28 40
(--) NVIDIA(0): 20 90 03 00 2c e1 10 00 00 18 00 00 00 00 00 00
(--) NVIDIA(0): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
(--) NVIDIA(0): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 d6
(--) NVIDIA(0): 20 00 c0 00 c3 71 b1 c6 26 cd 30 17 67 46 00 00
(--) NVIDIA(0): 01 10 01 00 3f 20 18 99 c8 04 88 a0 57 4a 9b 26
(--) NVIDIA(0): 3f c8 4c ff ff 80 81 80 81 8e 00 00 00 00 00 00
(--) NVIDIA(0): 00 00 00 00 00 00 e0 29 00 83 51 00 29 40 2a 68
(--) NVIDIA(0): 63 00 2c e1 10 00 00 04 bc 34 00 b0 51 00 28 40
(--) NVIDIA(0): 20 90 03 00 2c e1 10 00 00 18 00 00 00 00 00 00
(--) NVIDIA(0): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
(--) NVIDIA(0): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 d6
(--) NVIDIA(0):
(--) NVIDIA(0): --- End of EDID for P@@ &?0?gF (CRT-0) ---

Needless to say that all modes for CRT-0 are rejected due to the
EDID maximum pixel clock : 0.0 MHz

It does:

(WW) NVIDIA(0): Mode is rejected: PixelClock (31.5 MHz) too high for EDID
(WW) NVIDIA(0): (EDID Max: 0.0 MHz).

For everything.

So, what I need - apart from a new driver release that fixes this - is a way to
COMPLETELY disable EDID. I already added the (new) EDID options:

Option "UseEdidFreqs" "false"
Option "ModeValidation" "NoEdidModes"
Option "UseEDIDDpi" "false"
Option "DPI" "80 x 81"

But then it still uses the EDID PixelClock of 0 Mhz...
(some sanity check is missing there, to say the least).

Note that the Raw EDID bytes are the same. It's a bug in the decoding thereof.

Carlo

alaric 07-01-06 06:08 PM

Re: 1.0-8762 results in "black screen", 1.0-7676 does not. Using FX5200.
 
Ok, I now use:

# EDID is broken at the moment.
Option "UseEdidFreqs" "false"
Option "ModeValidation" "NoEdidMaxPClkCheck, NoEdidModes"
Option "UseEDIDDpi" "false"
Option "DPI" "80 x 81"


Adding the 'NoEdidMaxPClkCheck' solved the last problem,
it now works as before.

Still should be fixed of course.

Carlo

alaric 07-01-06 07:58 PM

Re: 1.0-8762 results in "black screen", 1.0-7676 does not. Using FX5200.
 
2 Attachment(s)
ARGH! It doesn't work anymore...

After I had a working xorg.conf that worked around the EDID processing bug,
I did the following:

1) Added agp=0 to boot parameters
2) Set NvAgp option to 3
3) Rebooted kernel 2.6.17 (most of the previous was with 2.6.14)

result: two black monitors (see attached log)

4) Removed agp=0 again
5) Set NvAgp option to 0 again
6) Rebooted kernel 2.6.14 again ...

result: two black monitors (!) (see second log)
THIS USED TO WORK A MINUTE AGO!?

Things that are different now are: No EDID is read at all anymore.
I see WAIT messages ... there is an Xid ... HELP!

I did a power off, just in case the video card was in a bad state that wasn't
reset over a normal reboot, and/or the BIOS(?) ... but that had no effect at all.

I had to reinstall 1.0-7676 again to get X working.

alaric 07-01-06 08:18 PM

Re: 1.0-8762 results in "black screen", 1.0-7676 does not. Using FX5200.
 
1 Attachment(s)
Story continuous:

After that last attempt (nvidia-bug-report-2.6.14-after-pwr-off.log.zip)
I did reboot (still 2.6.14) reinstalled 1.0-7676, started X (worked),
posted the above report and then:

- left X (contrl-alt-backspace)
- reinstalled 1.0-8762
- started X again

It works again (see attached log)

Conclusion: Running the 1.0-7676 driver does reset/fix something that
is permanently broken after trying 2.6.17 and/or agp=0 and/or Option "NvAGP" "3".
Something that isn't even reset over a power off.

The first thing I'm going to try now is rebooting to 2.6.17 without changing
the agp config (I NEED that kernel version) :/

alaric 07-01-06 08:50 PM

Re: 1.0-8762 results in "black screen", 1.0-7676 does not. Using FX5200.
 
1 Attachment(s)
Ok, I just rebooted to 2.6.17 (still having NvAGP = 0) and it hung again.
Going back to 2.6.14 still hung. I reinstalled 7676 again to recover.

Conclusion: it is impossible for me to run 2.6.17 :(
I can't use 'nv' because I need to TwinView :/.

Log file attached of the attempt to boot to 2.6.17

Carlo

alaric 08-08-06 08:32 AM

Why is nobody helping me?
 
Why is nobody helping me? I've waited a month now.

I am STILL not able to run kernel 2.6.17

mo_x 08-08-06 12:22 PM

Re: Why is nobody helping me?
 
Did you reinstall the driver after switching to the 2.6.17 kernel?

alaric 08-09-06 06:16 AM

Re: 1.0-8762 results in "black screen", 1.0-7676 does not. Using FX5200.
 
Yes - I always recompile/reinstall the driver after changing kernels.
This is hard to forget, since if I don't, then I can't start X -- well, in
this case I can't start it anyway, but yes, I did.

legluondunet 08-12-06 10:06 AM

Re: 1.0-8762 results in "black screen", 1.0-7676 does not. Using FX5200.
 
Hello,

I had a problem similar as yours.
I solved it with this two lines in xorg.conf:

Section "Device"
Identifier "NVIDIA Corporation NV35 [GeForce FX 5900XT]"
Driver "nvidia"
Option "RandRRotation" "true"
Option "IgnoreDisplayDevices" "CRT,TV"
EndSection

Section "Screen"
Option "backingstore" "true"
Identifier "Default Screen"
Device "NVIDIA Corporation NV35 [GeForce FX 5900XT]"
Monitor "Samsung SyncMaster 205bw"
DefaultDepth 24
Option "ConnectedMonitor" "DFP"
Option "NoLogo" "1"
Option "NvAGP" "1"

I hope this help you too :D

LGDN.

chris_z 08-14-06 02:28 AM

Re: 1.0-8762 results in "black screen", 1.0-7676 does not. Using FX5200.
 
Hi.
I had 'black screen' problem too, and I solved it by removing the following lines from xorg.conf
Option "IgnoreDisplayDevices" "DFP,TV"
Option "ConnectedMonitor" "CRT"

I hope it helps.

Cheers, Christopher

alaric 08-14-06 08:10 PM

Re: 1.0-8762 results in "black screen", 1.0-7676 does not. Using FX5200.
 
Yeah, well. I don't think this is related to my problem.


All times are GMT -5. The time now is 04:08 PM.

Powered by vBulletin® Version 3.7.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright 1998 - 2014, nV News.